A hybrid airspace model for utilizing drones in logistics operations

Abstract This study evaluates the effectiveness of the Biased Random-Key Genetic Algorithm (BRKGA) in solving the Traveling Salesman Problem (TSP) and its extension with trucks and drones, the Flying Sidekick Traveling Salesman Problem (FSTSP). The algorithm was tested on benchmark instances and two real-world scenarios. BRKGA achieved an average gap of 1.4% for TSP and showed better runtime compared to Simulated Annealing (SA) for smaller FSTSP instances. Although it produced faster computational results than SA for medium and large instances, the solution quality was slightly lower. When compared to leading metaheuristics like HTGVNS and HGVNS, BRKGA had competitive runtimes but was less optimal in solutions. In practice, the algorithm reduced delivery times and enabled drones to reach difficult locations. These results suggest that BRKGA is a valuable and effective heuristic for routing problems involving hybrid truck-drone systems.